Low frequency electrical stimulation through subdural electrodes in a case of refractory status epilepticus

Article (PDF Available)inClinical Neurophysiology 117(4):781-8 · May 2006with15 Reads
DOI: 10.1016/j.clinph.2005.12.010 · Source: PubMed
Abstract
We delivered low frequency stimulation through subdural electrodes to suppress seizures in a case of refractory status epilepticus (RSE). A 26-year-old female developed RSE after several days of febrile illness. Seizure control required continuous infusion of two anesthetics plus high doses of 2-4 enteral antiepileptic drugs. After 3 months of RSE, subdural strips were placed to determine surgical candidacy. Five independent ictal onset zones were identified. Because she was a poor candidate for epilepsy surgery and had a poor prognosis, the implanted subdural electrodes were used to administer 0.5 Hz stimulations to the ictal onset zones in 30 min trains daily for 7 consecutive days in an attempt to suppress seizures. After 1 day of stimulation, one anesthetic agent was successfully discontinued. Seizures only returned by the 4th day when the second anesthetic had been reduced by 60%. Upon returning, seizures arose from only one of the 5 original ictal onset zones. Unfortunately, RSE persisted, and she eventually died. In this case of RSE, low frequency stimulation through subdural electrodes transiently suppressed seizures from all but one ictal onset zone and allowed significant reduction in seizure medication. Low frequency cortical stimulation may be useful in suppressing seizures.
Low frequency electrical stimulation through subdural electrodes in a case
of refractory status epilepticus
Lara M. Schrader
a,
*
, John M. Stern
a
, Charles L. Wilson
a
, Tony A. Fields
a
, Noriko Salamon
b
,
Marc R. Nuwer
a
, Paul M. Vespa
c,a
, Itzhak Fried
c
a
Department of Neurology, David Geffen School of Medicine at UCLA, 710 Westwood Plaza, Room 1-194, Los Angeles, CA 90095, USA
b
Department of Radiologic Science, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095, USA
c
Department of Neurosurgery, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095, USA
Accepted 3 December 2005
Available online 2 February 2006
Abstract
Objective: We delivered low frequency stimulation through subdural electrodes to suppress seizures in a case of refractory status epilepticus
(RSE).
Methods: A 26-year-old female developed RSE after several days of febrile illness. Seizure control required continuous infusion of two
anesthetics plus high doses of 2–4 enteral antiepileptic drugs. After 3 months of RSE, subdural strips were placed to determine surgical
candidacy. Five independent ictal onset zones were identified. Because she was a poor candidate for epilepsy surgery and had a poor
prognosis, the implanted subdural electrodes were used to administer 0.5 Hz stimulations to the ictal onset zones in 30 min trains daily for 7
consecutive days in an attempt to suppress seizures.
Results: After 1 day of stimulation, one anesthetic agent was successfully discontinued. Seizures only returned by the 4th day when the
second anesthetic had been reduced by 60%. Upon returning, seizures arose from only one of the 5 original ictal onset zones. Unfortunately,
RSE persisted, and she eventually died.
Conclusions: In this case of RSE, low frequency stimulation through subdural electrodes transiently suppressed seizures from all but one ictal
onset zone and allowed significant reduction in seizure medication.
Significance: Low frequency cortical stimulation may be useful in suppressing seizures.
q 2006 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
Keywords: Low frequency repetitive stimulation; Refractory status epilepticus; Treatment; Electrical stimulation; Cortical stimulation; Seizures
1. Introduction
Refractory status epilepticus (RSE), defined as the
persistence of status epilepticus despite treatment with
intravenous medications, occurs in approximately 30% of
patients with status epilepticus (Mayer et al., 2002) and is
associated with high morbidity and mortality (Claassen
et al., 2002). Treatment with surgical resection of
epileptogenic tissue has been reported (Gorman et al.,
1992; Ma et al., 2001; Van Ness, 1990) but is not an option
in many cases. Thus, there is a clear need for therapeutic
alternatives.
Animal studies have shown that low frequency repetitive
electrical stimulation (LFRES) of the cerebral cortex may
be therapeutic for seizures (Ullal et al., 1989). There is also
evidence that low frequency repetitive transcranial mag-
netic stimulation (LF-rTMS), which produces indirect
electrical stimulation of the cortex, may be therapeutic in
epilepsy (Akamatsu et al., 2001; Daniele et al., 2003; Fregni
et al., 2005; Menkes and Gruenthal, 2000; Tergau et al.,
1999; Theodore et al., 2002).
Because of this, we administered LFRES directly
through subdural electrodes in an individual with RSE
who was a poor candidate for resective epilepsy surgery and
Clinical Neurophysiology 117 (2006) 781–788
www.elsevier.com/locate/clinph
1388-2457/$30.00 q 2006 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.clinph.2005.12.010
*
Corresponding author. Tel.: C1 310 206 3093; fax: C1 310 267 1157.
E-mail address: lschrader@mednet.ucla.edu (L.M. Schrader).
who had been refractory to 3 months of high-dose
combination antiepileptic drug (AED) therapy. Because of
her poor prognosis and lack of treatment alternatives, a
decision was made to consider an investigational interven-
tion, LFRES, as compassionate care. This was possible to
offer with low risk because the patient had already had
subdural electrodes placed as part of her epilepsy surgery
evaluation.
2. Case report and methods
2.1. Case
A previously healthy 26-year-old female was admitted
after 1 day of confusion and 6 days of fever, cough and
emesis. She had two generalized tonic–clonic seizures on
the day of admission and 4 on the following day. Head CT,
CSF analysis, and brain MRI/MRS upon presentation were
unremarkable. Six days after admission, she developed
successive seizures, ultimately requiring intubation and
continuous pentobarbital infusion due to the development of
status epilepticus. The EEG was kept in burst-suppression
for 3 months. Seizure control required continuous intrave-
nous infusion with two anesthetics in addition to high doses
of 2–4 enteral AEDs. Over the 3 months, numerous attempts
were made to wean off anesthetic agents. When one
anesthetic agent was weaned, seizure activity occurred
within hours and scalp EEG monitoring demonstrated (1)
interictal discharges independently occurring at the T5, F3,
and T3 electrodes, (2) bilateral periodic epileptiform
discharges (BiPEDs) associated with right jaw and right
neck twitching, and (3) left occipital electrographic seizures
occurring within hours and associated with eye blinking.
The two anesthetic agents that were used at any one time
were periodically changed to determine if better seizure
control could be achieved with different combinations of
agents. At any one time, a combination of two of 3
anesthetic agents, propofol, pentobarbital or midazolam,
was administered continuously with doses titrated to
achieve burst-suppression. For similar reasons, the combi-
nation of enteral AEDs likewise varied over time, and the
following AEDs were used at some time in 2- to 4-AED
combination therapy: phenytoin 600–1550 mg per day
(blood levels typically ranged from 15 to 30 mcg/mL),
levetiracetam 3000–7500 mg per day, lamotrigine 100–
250 mg per day, felbamate 4500–7500 mg per day,
topiramate 400 mg per day, valproate 7500–9500 mg per
day (blood levels typically ranged from 50 to 100 mcg/mL),
phenobarbital 460–600 mg per day (blood levels typically
ranged from 40 to 100 mcg/mL), and clonazepam 6 mg per
day. Partial pressures of arterial O
2
and CO
2
(PaO
2
and
PaCO
2
) were kept within normal limits, and there was no
correlation between relatively minor changes in arterial pH
and incidence of seizures.
Multiple neuroimaging studies were done throughout the
hospitalization. A brain MRI 1 month after admission was
unremarkable. A repeat MRI 2 months after admission
showed diffuse increased signal on T2 and FLAIR of both
gray and white matter as well as bilateral decrease in
hippocampal size in comparison to the previous MRI.
Another MRI 3 months after admission revealed a mild
Fig. 1. Axial and left hemisphere sagittal views of an ictal fluorodeoxyglucose positron emission tomography (FDG-PET) scan superimposed upon MRI
demonstrate marked increased activity in the left medial occipital region with patchy areas of increased activity in the left superior parietal and right medial
occipital lobes.
L.M. Schrader et al. / Clinical Neurophysiology 117 (2006) 781–788782
Fig. 2. (A–E) Three-dimensional surface rendering MRI reconstruction demonstrates the location of subdural electrode strips, labeled as follows: FAL, frontal
anterior lateral; FAT, frontal anterior temporal; FMT, frontal mid-temporal; FTP, frontal temporal parietal; PF, parietal frontal; PT, parietal temporal; POL,
parietal occipital (lateral), POM, parietal occipital (medial); POML, parietal occipital (mid-line); RPF, right parietal frontal; RPT, right parietal temporal; RPO,
right parietal occipital. POML was a mid-line strip that cannot be seen in the views shown here. Each strip had 4–12 electrode contacts, numbered
consecutively. Each strip is demonstrated using a different color. Colored numbers indicate the numbering of the corresponding color-matched strip. A white
color denotes electrodes that were most consistently involved in the ictal onset pattern. The following patterns are shown: (2A) left temporoparietal, (2B) right
frontal, (2C) left parieto-occipital, (2D) left parietal, and (2E) right occipital.
L.M. Schrader et al. / Clinical Neurophysiology 117 (2006) 781–788 783
increase in atrophy and less prominent generalized
increased signal on T2 and FLAIR images. Two months
after admission, an FDG-PET during a BiPED EEG pattern
revealed marked hypermetabolism in the left occipital lobe
with patchy areas of hypermetabolism in the left parietal and
right occipital lobes. An ictal FDG-PET 3 months after
admission revealed marked activity in the left occipital lobe
with patchy areas of increased activity in the left parietal
and right occipital lobes (Fig. 1).
Because of there was a suggestion of a possible
epileptogenic region on scalp EEG monitoring and FDG-
PET studies and because the patient had had an unsatisfac-
tory response to AEDs, a decision was made to proceed with
an evaluation for epilepsy surgery. A methohexital
suppression test demonstrated seizure suppression with a
left but not a right internal carotid artery injection. In all,
these results supported the possibility of a left posterior
epileptogenic region. To further assess epilepsy surgery
candidacy, subdural electrode strips with 108 contacts were
placed across bilateral hemispheres with greater coverage
on the left. Five independent ictal onset zones were
identified (Figs. 2 and 3): (A) left temporoparietal (27% of
Fig. 3. (A–E) A bipolar montage is used to demonstrate the 5 ictal onset patterns using 1 Hz low frequency and 70 Hz high frequency filter settings. Sensitivity
settings are 150 mV Fig. 3A and B and 300 mV for Fig. 3C–E. Electrode labels are defined in the legend for Fig. 2. The spatial orientation of electrodes is
demonstrated in Fig. 2. The 5 ictal patterns are: (3A) left temporoparietal, (3B) right frontal, (3C) left parieto-occipital, (3D) left parietal, and (3E) right
occipital.
L.M. Schrader et al. / Clinical Neurophysiology 117 (2006) 781–788784
seizures), (B) right frontal (3%) (C) left parieto-occipital
(14%), (D) left parietal (12%), and (E) right occipital (44%).
Seizure type C was associated with blinking, and the other
ictal patterns were not accompanied by behavioral change.
During a 1 h time sample while midazolam 6 mg/h and
propofol 59 mcg/kg were administered, 15 seizures were
recorded. Because multiple independent ictal onset zones
were present, the patient was considered a poor candidate
for epilepsy surgery.
2.2. Stimulation protocol
Given the poor surgical candidacy and unsatisfactory
response to 3 months of multiple AEDs, the prognosis was
becoming progressively unfavorable. Because of the poor
outlook and a lack of treatment alternatives, a decision was
made to consider an investigational intervention, LFRES, as
compassionate care. This was possible to offer with low risk
because the patient had already had subdural electrodes
placed as part of her epilepsy surgery evaluation. Because of
technical limitations, 8 was the maximum number of
electrode pairs that could be stimulated at one time. We
were therefore, unable to deliver stimulations to all ictal
onset zones simultaneously and so stimulated, different
brain regions consecutively each day. For each brain region
stimulated, 4–8 neighboring electrode pairs were stimulated
at a time. Two to 4 consecutive 30 min sessions separated by
10–15 min were applied to different brain regions each day
for 7 consecutive days. Square wave biphasic pulse
stimulations of 500 ms per phase at 0.5 Hz were adminis-
tered to the ictal onset zones through adjacent pairs of
electrodes using a Grass S12 Isolated Biphasic Stimulator
(Grass Instrument Co., Quincy, MA, USA). Stimulations
were delivered in parallel simultaneously to 4–8 electrode
pairs. Table 1 shows the electrodes pairs that were
stimulated each day for each session. The stimulation
pairs chosen each day were based on clinical judgment
regarding which brain regions appeared to be most involved
in seizure generation. As seen in Table 1, the ictal onset
zones with the highest seizure frequency were targeted with
stimulation on the first 2 days, and more electrode pairs
were added over subsequent days in an attempt to decrease
cortical excitability over all ictal onset zones. On the 4th and
7th day of stimulation, one of the 30 min stimulation trains
was interrupted early because of the occurrence of a seizure
that electrographically resembled spontaneous seizures that
were seen earlier that day. After 7 days, the electrodes were
removed to minimize risks of infection.
For all stimulations, the stimulator intensity was 16 mA.
This resulted in 2–4 mA total current per electrode pair and
charge densities of 14.2–28.3 mC/cm
2
per phase, depending
upon the number of electrode pairs stimulated
Table 1
Pairs of electrodes stimulated each day
Day Session Electrode pairs through, which stimulation was administered
1 1 FTP 7(K)-8(C)&9(C)-10(K); PF 5(K)-6(C); FMT 5(K)-6(C)
2 FTP 1(C)-2(K); POML 1(C)-2(K)&3(K)-4(C); PT 8(C)-9(K)
3 RPO 1(C)-2(K)&3(K)-4(C); RPF 1(C)-2(K)&3(K)-4(C); RPT 1(C)-2(K)&3(K)-4(C)
2 1 FTP 7(K)-8(C)&9(C)-10(K); PF 5(K)-6(C), FMT 5(K)-6(C); FTP 1(C)-2(K); POML 1(C)-2(K)&3(K)-4(C); PT 8(C)-9(K)
2 RPO 1(C)-2(K)&3(K)-4(C); RPF 1(C)-2(K)&3(K)-4(C); RPT 1(C)-2(K)&3(K)-4(C)
3 1 FTP 7(K)-8(C)&9(C)-10(K); PF 5(K)-6(C); FMT 3(C)-4(K) &5(K)-6(C)
2 FTP 1(C)-2(K); POML 1(C)-2(K)&3(K)-4(C); PT 8(C)-9(K)
3 RPO 1(C)-2(K)&3(K)-4(C); RPF 1(C)-2(K)&3(K)-4(C); RPT 1(C)-2(K)&3(K)-4(C)
4 1 FTP 7(K)-8(C)&9(C)-10(K); PF 5(K)-6(C); FMT 3(C)-4(K)&5(K)-6(C)
2 FTP 1(C)-2(K); POML 1(C)-2(K)&3(K)-4(C); PT 8(C)-9(K); POM 5(K)-6(C)
3 POML 1(C)-2(K)&3(K)-4(C); POM 1(C)-2(K)&3(K)-4(C)&5(C)-6(K); POL 1(C)-2(K)&3(K)-4(C)&5(C)-6(K)
4 RPO 1(C)-2(K)&3(K)-4(C); RPF 1(C)-2(K)&3(K)-4(C); RPT 1(C)-2(K)&3(K)-4(C)
5 1 POML 1(C)-2(K)&3(K)-4(C); POM 1(C)-2(K)&3(K)-4(C)&5(C)-6(K); POL 1(C)-2(K)&3(K)-4(C)&5(C)-6(K)
2 FTP 7(K)-8(C)&9(C)-10(K); PF 5(K)-6(C); FMT 3(C)-4(K)&5(K)-6(C); PT 2(K)-3(C) &8(C)-9(K)
3 RPO 1(C)-2(K)&3(K)-4(C); RPF 1(C)-2(K)&3(K)-4(C); RPT 1(C)-2(K)&3(K)-4(C)
6 1 POML 1(C)-2(K)&3(K)-4(C); POM 1(C)-2(K)&3(K)-4(C)&5(C)-6(K); POL 1(C)-2(K)&3(K)-4(C)&5(C)-6(K)
2 FAT 1(C)-2(K)&3(K)-4(C)&5(C)-6(K)&7(K)-8(C); FTP 1(C)-2(K)&3(K)-4(C); FAL 1(C)-2(K)&3(K)-4(C)
3 FTP 7(K)-8(C)&9(C)-10(K); PF 5(K)-6(C); FMT 3(C)-4(K)&5(K)-6(C)PT2(K)-3(C)&4(C)-5(K) &8(C)-9(K)
4 RPO 1(C)-2(K)&3(K)-4(C); RPF 1(C)-2(K)&3(K)-4(C); RPT 1(C)-2(K)&3(K)-4(C)
7 1 FAT 1(C)-2(K)&3(K)-4(C)&5(C)-6(K)&7(K)-8(C); FTP 1(C)-2(K)&3(K)-4(C)& 5(C)-6(K); FAL 1(C)-2(K)
2 POML 1(C)-2(K)&3(K)-4(C); POM 1(C)-2(K)&3(K)-4(C)&5(C)-6(K); POL 1(C)-2(K)&3(K)-4(C)&5(C)-6(K)
3 FTP 7(K)-8(C)&9(C)-10(K); PF 5(K)-6(C); FMT 3(C)-4(K)&5(K)-6(C)PT2(K)-3(C)&4(C)-5(K) &8(C)-9(K)
4 RPO 1(C)-2(K)&3(K)-4(C); RPF 1(C)-2(K)&3(K)-4(C); RPT 1(C)-2(K)&3(K)-4(C)
This table shows the pairs of electrodes that were stimulated at each session each day and the electrodes that were the anode (K) and cathode (C) of each pair.
The electrodes that were stimulated on Day 1 continued to be stimulated on subsequent days. If a new electrode pair was added on a subsequent day, it appears
in bold. The following abbreviations refer to the name of each strip: FAL, frontal anterior lateral; FAT, frontal anterior temporal; FMT, frontal mid-temporal;
FTP, frontal temporal parietal; PF, parietal frontal; PT, parietal temporal; POL, parietal occipital (lateral); POM, parietal occipital (medial); POML, parietal
occipital (mid-line); RPF, right parietal frontal; RPT, right parietal temporal; RPO, right parietal occipital. Each strip had 4–12 electrode contacts, numbered
consecutively. The numbers listed in the electrode pairs column represent the number assigned to the electrode of that subdural strip.
L.M. Schrader et al. / Clinical Neurophysiology 117 (2006) 781–788 785
simultaneously, which ranged from 4 to 8. The decision as
to whether to use 4, 6 or 8 electrode pairs per stimulation
session was based on clinical judgment about the extent of
cortical surface to be stimulated each day and time
considerations.
3. Results
The patient was seizure-free on midazolam 10 mg/h and
propofol 51 mcg/kg per minute. The propofol was reduced
from 51 to 30 mcg/kg per minute 2 h prior to starting LFRES.
Seizures did not recur during these 2 h. After 1 day of LFRES,
propofol was successfully discontinued without recurrence of
seizures. This was considered an improvement since in the
preceding 3 months attempts to wean to only one anesthetic
agent resulted in breakthrough seizures within hours. Seizures
remained absent for the following 3 days, despite a slow wean
of midazolam over this time. Midazolam was reduced to
8 mg/h the morning of the second day of LFRES and to 6 mg/h
the morning of the 3rd day of LFRES without any seizure
recurrence. This seizure freedom while off of propofol and on
doses as low as 6 mg/h of midazolam is contrasted to the
preceding week when seizures occurred an average of 7 times
per hour while on a steady dose of propofol 22 mcg/kg per
minute and midazolam 6 mg/h. During the first 3 days of
LFRES, the interictal electrocorticography (ECoG) showed
intermittent left-sided periodic epileptiform discharges that
generally occurred when the patient was physically stimulated
by staff.
Interestingly, the LFRES also changed the interictal ECoG.
In the hours preceding the first day’s LFRES sessions, the
ECoG background showed sustained w1 Hz left-sided
periodic epileptiform discharges in left posterior quadrant
(Fig. 4(A)). Immediately after LFRES there was a noticeable
decrease in frequency of these discharges, and within 1 h the
discharges were not only less frequent but also had a less broad
cortical distribution (Fig. 4(B)).
Seizures returned on the 4th day when midazolam had been
reduced by 60% to 4 mg/h. Upon returning, Pattern C was the
only ictal onset pattern, and its electrophysiological appear-
ance was no different in frequency or duration from the pattern
that appeared prior to starting LFRES. The other 4 seizure
patterns did not recur on this day or on the subsequent days of
LFRES. Despite the recurrence of seizure Pattern C on the 4th
day, the treating physician continued tapering midazolam to
see if the patient might regain awareness since the recurring
seizures were highly focal. On the 4th day, the midazolam was
tapered off by noon. The patient was on midazolam 1 ml/h for
4 h and then off of midazolam for another 4 h. The patient did
not regain awareness, and Pattern C seizures continued.
Therefore, the midazolam was increased to 4 mg/h. At this
dose, Pattern C seizures continued at a frequency of about 6 per
hour. Because of this continued seizure activity, on the 5th day
midazolam was increased to 8 mg/h and then 16 mg/h. She
remained on this midazolam dose and off of propofol through
the seventh day of LFRES and continued to have only type C
seizures at a frequency of about 6 per hour. Because the patient
continued to have seizures and did not regain awareness with
medication taper, a decision was made to perform resection of
the epileptogenic region most resistant to suppression, which
was also the region that was also most active on PET.
Subsequently, the electrode strips were removed,
followed by left parieto-occipital resection and selective
multiple subpial transection. Pathology study of the surgical
specimen revealed prominent reactive changes in the
cerebral cortex and white matter. There was also evidence
suggestive of acute or subacute global anoxia/ischemia.
Mild focal Chaslin’s gliosis was present at the pial surface.
The seizures continued after surgery. Similar to her needs
prior to LFRES, seizure control again required high doses of
two anesthetic agents, midazolam and propofol, for seizure
control. The localization of persisting seizures was unclear
because subdural electrodes had been removed. Because of her
poor prognosis, the patient was transferred to a subacute
hospital. She died 2 months later, 5 months after illness onset.
Changes noted at autopsy were thought to be secondary to
prolonged seizure activity and anoxic/ischemic injury, and no
cause for her seizures was found.
Fig. 4. (A and B) A bipolar montage is used to demonstrate interictal
patterns using 1 Hz low frequency and 70 Hz high frequency filter settings.
The sensitivity setting is 300 (V. Electrode labels are defined in the legend
for Fig. 2. The spatial orientation of these electrodes is demonstrated in
Fig. 2. In the hours prior to the first day’s LFRES sessions, sustained w1Hz
broadly distributed left-sided periodic epileptiform discharges were seen
(4A). An hour after completion of the first day’s LFRES sessions, the
discharges became less frequent and less broadly distributed (4B).
L.M. Schrader et al. / Clinical Neurophysiology 117 (2006) 781–788786
4. Discussion
In this case of RSE, LFRES through subdural strips
transiently suppressed seizures from all but one ictal onset
zone and allowed a significant reduction in seizure
medication. Because medications were not held constant
during the period of stimulation nor postoperatively, it is
difficult to accurately assess the degree or duration of
antiepileptic effect. While her underlying illness was
ultimately catastrophic, the transient reduction in seizure
activity after LFRES substantiates the potential for such an
intervention to be efficacious in treating epilepsy.
While it is possible that the absence of seizures during the
first 3 days of LFRES was coincidental and not caused by
LFRES, this is unlikely based on the data from the prior 3
months of monitoring. During this time, attempts to wean to
only one anesthetic agent resulted in breakthrough seizures
within hours. Furthermore, after LFRES no seizures occurred
for 3 days on midazolam alone, while in contrast, 1 week prior
to LFRES frequent seizures were observed with comparable
doses of midazolam plus high-dose propofol.
The precise mechanism by which low frequency stimu-
lation may reduce seizures is unknown. It may exert its
therapeutic effect by a mechanism similar to that by which
long-term depression is produced from LFRES. Long-term
depression is a partially N-methyl-
D-aspartate (NMDA)-
dependent form of synaptic plasticity that results in long-
lasting decreased synaptic responsiveness (Lei and McBain,
2004; Manabe, 1997). This long-lasting decreased synaptic
responsiveness may be responsible for a therapeutic effect of
low frequency cortical stimulation in epilepsy.
It is interesting that seizures were suppressed for some
but not all ictal onset zones. The ictal onset zone that was
most difficult to suppress in this case was pattern C, which
arose from the left parieto-occipital region. It is possible that
this region might have been pathophysiologically different
than the other regions, and thus perhaps NMDA-mediated
influences on synaptic plasticity may not be efficacious for
all seizure types. Another potential reason for the difficulty
in suppressing this epileptogenic region is that it may have
been larger and deeper than the other epileptogenic regions.
This epileptogenic region was also the most hypermetabolic
on the ictal FDG-PET, as seen in Fig. 1. This figure
demonstrates that this ictal region included not only the
brain tissue along the lateral convexity where some
electrodes were located but also included the medial
parasagittal tissue that did not have overlying electrodes.
Thus, perhaps such a large epileptogenic region may be
difficult to suppress through stimulation of only a few
electrodes lying over a small portion of it, and perhaps
stimulation through broader electrode coverage would have
made full seizure suppression possible.
Animal studies have shown that low frequency repetitive
stimulation of the cortex may be therapeutic for seizures.
Low frequency repetitive electrical stimulation increases
after-discharge thresholds during the course of hippocampal
and amygdaloid kindling (Ullal et al., 1989). Another rat
study demonstrated that LF-rTMS directed at the parietal
lobe prior to the administration of pentylenetetrazol (PTZ)
delayed the onset of PTZ-induced seizures. Furthermore,
100% of control rats developed status epilepticus after PTZ
injection compared to 40% of rats pretreated with LF-rTMS
(Akamatsu et al., 2001).
The use of LF-rTMS to treat people with epilepsy has
shown promise. In a person with seizures originating from a
focal cortical dysplasia, seizure frequency was reduced by
70% after treatment with LF-rTMS directed at the epileptic
focus (Menkes and Gruenthal, 2000). In a study of 9
subjects, LF-rTMS directed at the vertex resulted in 3
patients with a 20–50% reduction in seizure frequency and 3
with a O50% reduction (Tergau et al., 1999). In a third
study of 24 patients with localization-related epilepsy
randomized to either sham or LF-rTMS directed at the
epileptogenic cortex (Theodore et al., 2002), subjects
receiving LF-rTMS had reduced seizure frequency for 2
weeks after LF-rTMS. These results did not reach statistical
significance, perhaps due to the study’s low power. In this
third study, patients with neocortical versus mesial temporal
epileptogenic foci tended to have a more robust decrease in
seizure frequency, perhaps due to the deeper location of
mesial temporal structures and thus increased distance from
the TMS coil. A 4th study of 4 subjects with intractable
seizures demonstrated reduced seizure frequency when LF-
rTMS was directed at the epileptic focus in individuals with
a single epileptic focus but no reduction of seizure
frequency when LF-rTMS was directed at the vertex in
individuals with multifocal epilepsy (NZ2) (Daniele et al.,
2003). Lastly, a 5th study of 8 patients with malformations
of cortical development who underwent LF-rTMS stereo-
tactically targeted at their dysplasia experienced a signifi-
cant reduction in epileptiform discharge and seizure
frequency (Fregni et al., 2005).
In the present case of RSE, LFRES through subdural
electrodes was accompanied by transient suppression of all
but one epileptic focus, allowing significant temporary
reduction in seizure medication. Unfortunately, the final
outcome was poor, as it is in many cases of this nature.
Nevertheless, this case provides evidence that low frequency
cortical stimulation may be an efficacious treatment for
epilepsy.
References
Akamatsu N, Fueta Y, Endo Y, Matsunaga K, Uozumi T, Tsuji S.
Decreased susceptibility to pentylenetetrazol-induced seizures after
low-frequency transcranial magnetic stimulation in rats. Neurosci Lett
2001;310:153–6.
Claassen J, Hirsch LJ, Emerson RG, Mayer SA. Treatment of refractory
status epilepticus with pentobarbital, propofol, or midazolam: a
systematic review. Epilepsia 2002;43:146–53.
L.M. Schrader et al. / Clinical Neurophysiology 117 (2006) 781–788 787
Daniele O, Brighina F, Piazza A, Giglia G, Scalia S, Fierro B. Low-
frequency transcranial magnetic stimulation inpatients with cortical
dysplasia: a preliminary study. J Neurol 2003;250:761–2.
Fregni F, Thome-Souza S, Bermpohl F, Marcolin MA, Herzog A,
Pascual-Leone A, Valente KD. Antiepileptic effects of repetitive
transcranial magnetic stimulation in patients with cortical malfor-
mations: an EEG and clinical study. Stereotact Funct Neurosurg
2005;83:57–62.
Gorman DG, Shields WD, Shewmon DA, Chugani HT, Finkel R,
Comair YG, Peacock WJ. Neurosurgical treatment of refractory status
epilepticus. Epilepsia 1992;33:546–9.
Lei S, McBain CJ. Two loci of expression for long-term depression at
hippocampal mossy fiber-interneuron synapses. J Neurosci 2004;24(9):
2112–21.
Ma X, Liporace J, O’Connor MJ, Sperling MR. Neurosurgical treatment
of medically intractable status epilepticus. Epilepsy Res 2001;46:
33–8.
Manabe T. Two forms of hippocampal long-term depression, the
counterpart of long-term potentiation. Rev Neurosci 1997;8:179–93.
Mayer SA, Claassen J, Lokin J, Mendelsohn F, Dennis LF, Fitzsimmons B-
F. Refractory status epilepticus: frequency, risk factors, and impact on
outcome. Arch Neurol 2002;59:205–10.
Menkes DL, Gruenthal M. Slow-frequency repetitive transcranial magnetic
stimulation in a patient with focal cortical dysplasia. Epilepsia 2000;41:
240–2.
Tergau F, Naumann U, Paulus W, Steinhoff BJ. Low-frequency repetitive
transcranial magnetic stimulation improves intractable epilepsy. Lancet
1999;353:2209.
Theodore WH, Hunter K, Chen R, Vega-Bermudez F, Boroojerdi B,
Reeves-Tyer P, Werhahn K, Kelley KR, Cohen L. Transcranial
magnetic stimulation for the treatment of seizures: a controlled study.
Neurology 2002;59:560–2.
Ullal GR, Ninchoji T, Uemura K. Low frequency stimulation induces an
increase in after-discharge thresholds in hippocampal and amygdaloid
kindling. Epilepsy Res 1989;3:232–5.
Van Ness PC. Pentobarbital and EEG burst suppression in treatment of
status epilepticus refractory to benzodiazepines and phenytoin.
Epilepsia 1990;31:61–7.
L.M. Schrader et al. / Clinical Neurophysiology 117 (2006) 781–788788
    • "It is thought that the underlying mechanism is long term depression (LTD) that was investigated in hippocampus, visual cortex and motor cortex [19]. LTD is a form of synaptic plasticity and it results in long lasting decreased responsiveness [14]. The LTD induced by 1 Hz LFS in sensorimotor cortex depends on activation of NMDA glutamate receptors and voltage gated calcium chan- nels [21]. "
    [Show abstract] [Hide abstract] ABSTRACT: Low frequency electrical stimulation has been revealed that as a potential cure in patient with drug resistant to epilepsy. This study tries to evaluate the effect of low frequency electrical stimulation (LFS) on absence seizure of perioral region primary somatosensory cortex (S1po). Eighteen male WAG/Rij rats were received LFS (3Hz, square wave, monophasic, 200μs, and 400μA) for 25min into S1po for a period of five days. There is 6 animals per group .The stimulating electrodes were implanted according to stereotaxic landmarks and EEG recording was obtained 30min before and after LFS to analyse frequency, number and duration of spike-wave discharges (SWD). The results showed that in animals with unilateral stimulating electrodes (Exp1) in first and second days and also in animals with bilateral stimulating electrodes (Exp2) in days 3rd and 4th. LFS had significant decrease effects (p<0.05) on mean number of SWD between pre-LFS. In comparison pre-LFS to post-LFS, mean of duration in Exp2 decreased significantly. In continuous application of LFS (5 days) only the data of first day was differently significant (p<0.05) but data of other days had no difference. Comparison of data between Exp1, Exp2 and control groups showed that the mean number of Exp1 was significantly different (p<0.05) and mean pick frequency in Exp2 was significantly decreased in comparison with Exp1 group (p<0.05). The LFS of S1po produces significant antiepileptic effect on absence seizure but it was not persistent till the next day and shows a short time effect.
    Full-text · Article · Sep 2013
    • "One of the potential alternative therapies for epilepsy is deep brain stimulation(Kile, Tian, & Durand, 2010 ). Low frequency stimulation(LFS) as a form of deep brain stimulation August 2013, Volume 4, Number 3 is thought to inhibit the activity by increasing the threshold for the firing of neuronal action potentials through more complex mechanisms(Albensi, Ata, Schmidt, Waterman, & Janigro, 2004; Schrader et al., 2006 ). Moreover , LFS requires fewer pulses per second compared to other forms of deep brain stimulation therapies, thereby lowering the required current injection and minimizing the potential for the stimulation-induced damage of the target tissue(Kile et al., 2010). "
    [Show abstract] [Hide abstract] ABSTRACT: Low frequency stimulation (LFS) is a potential alternative therapy for epilepsy. However, it seems that the anticonvulsant effects of LFS depend on its target sites in the brain. Thus, the present study was designed to compare the anticonvulsant effects of LFS administered to amygdala, piriform cortex and substantia nigra on amygdala kindling acquisition. In control group, rats were kindled in a chronic manner (one stimulation per 24 h). In other experimental groups, animals received low-frequency stimulation (8 packages at 100 s intervals, each package contained 200 monophasic square-wave pulses, 0.1 ms pulse duration at 1 Hz andAD threshold intensity) in amygdala, piriform cortex or substantia nigra 60 seconds after the kindling stimulation, the AD duration and daily seizure stages were recorded. The obtained results showed that administration of LFS in all three regions reduced electrical and behavioral parameters of the kindling procedure. However LFS has a stronger inhibitory effect on kindling development when applied in substantia nigra compared to the amygdala and piriform cortex which reinforce the view that the substantia nigra mediates a crucial role in amygdala-kindled seizures. LFS had also greater inhibitory effects when applied to the amygdala compared to piriform cortex. Thus, it may be suggested that antiepileptogenic effect of LFS depends on its target site and different brain areas exert different inhibitory effects on kindling acquisition according to the seizure focus.
    Full-text · Article · Aug 2013
    • "High frequency stimulation (HFS) as a form of deep brain stimulation has been shown to inhibit pathologic neuronal activity through direct disruption of local network activities. In contrast low frequency stimulation (LFS) is thought to inhibit the activity by increasing the threshold for the firing of neuronal action potentials through more complex mechanisms (Albensi et al. 2004; Schrader et al. 2006). Moreover, LFS requires fewer pulses per second compared to other forms of DBS therapies, thereby lowering the required current injection and minimizing the potential for the stimulation-induced damage of the target tissue (Kile et al. 2010). "
    [Show abstract] [Hide abstract] ABSTRACT: In epilepsy, the anticonvulsant effects of low-frequency stimulation (LFS) are time dependent. We report the effects of 1.2g daily walnut (Juglans regia L) consumption after termination of kindling (Immediate LSF), and after cessation of after-discharges (Delayed LSF), on amygdaloid kindled seizures in male Wistar rats. Control and walnut consuming rats received daily kindling and low-frequency monophasic square-wave pulses of 1 Hz, 100 µA, 0.1 ms per pulse every 15 min. Results indicate an anti-convulsant effect of walnut pre-treatment in kindled induced seizures Walnut consumption and immediate LFS reduced electrical and behavioral parameters of kindling whereas delayed LFS had no significant effect. There was no significant interaction between the anticonvulsant effects of walnuts and LFS. Walnut consumption may have delayed the kindling procedure but did not interact with LFS effects.
    Full-text · Article · Jul 2013 · Basic and Clinical Neuroscience
Show more